DESCRIPTION: (Applicant's Description) The overall goal of this proposal is to combine extensive clinical experience with MRI/MRSI of cancer in the Department of Radiology, University of California San Francisco with the engineering expertise of General Electric Corporate Research and Development to develop and optimize technology for improved MRI/MRSI localization and staging of prostate cancer, and to perform MR-guided prostate biopsies on a clinical 1.5T MR scanner. We believe that it is appropriate to skip the exploratory/developmental phase grant (R21) and apply directly for the Exploratory/Developmental phase grant (R33) since the feasibility of the proposed MR techniques and the endorectal biopsy probe have already been demonstrated. The proposed grant addresses most of the specific features of the RFA including: 1) a Academic/Industry partnership, 2) improved MRI/MRSI technology that will lead to better prostate cancer staging and localization, 3) the implementation of "real-time" image processing methods for guiding and monitoring prostate biopsies, 4) improved tissue sampling through MRI/MRSI guided biopsies, 5) the design and implementation of an image-guided navigational system, and 6) the determination of clinical feasibility of performing MR guided biopsies. Additionally, the technology developed in this proposal for MR targeting of biopsies could also be used for targeting and monitoring of prostate cancer therapy in future studies. Accurate characterization of prostate cancers is a major problem facing the clinical management of individual prostate cancer patients and for monitoring clinical trials. Although transrectal ultrasound-guided biopsies are the current standard, they suffer from severe sampling errors since TRUS does not offer adequate accuracy in identifying the location and extent of prostate cancer. Over the past 7 years, we have developed 3D MRSI techniques which greatly improve localization of prostate cancer within the gland, improve the prediction of extracapsular spread of the cancer (ECE), and can assess cancer grade based on cellular metabolite levels. In this project, we propose to improve this powerful imaging technique by means of a higher sensitivity MR probe and develop techniques for MRI/MRSI guided biopsies. We believe that these developments will significantly improve the characterization of individual prostate cancers and will provide accurate tissue sampling which is critical for a large number of research projects and clinical trials.